Currently available noninvasive methods for measuring inspiratory resistance (RI) are difficult to implement or interpret during assisted mechanical ventilation on account of the confounding effect of respiratory efforts (Pmus). We propose a simple method consisting of brief reductions in airway pressure (Paw) in the early part of the inflation phase (pulse). Paw, flow (V), and volume (V) are measured at the beginning of the pulse (T (0)), at the trough of the pulse (TI) and at a point 0.1 s before T(0) (T(-1)). Equations of motion of the form [Pmus + Paw = V. K(1) + V (2). K(2) +V. E] are generated for the data at the three time points (E = elastance, K(1) and K(2) are Rohrer's constants). These three equations can be solved for K(1) and K(2) if it is arranged that the pulse has appropriate configuration and timing, and if it is assumed that DeltaPmus/Deltat is constant over the brief pulse period. The method was tested in 67 patients ventilated with proportional assist ventilation (PAV). The results were compared with those obtained using the interrupter technique during a period of controlled mechanical ventilation (CMV). RI, expressed at a standard flow of 1 L. s(-)(1), was slightly higher during PAV (16.4 +/- 4.9 versus 15.5 +/- 4.5 cm H(2)O. L(-1). s, p < 0.001). The average difference was 0.9 +/- 2.0 cm H(2)O. L(-1). s, corresponding to 5.4 +/- 12.6% of the average of RCMV and RPAV. The correlation coefficient was 0.92 (p = 8E-28) with a slope (1.01) and intercept (0.8) not significantly different from 1.0 and 0, respectively. We conclude that brief negative pulses applied early during the inflation phase can be used to provide reliable estimates of inspiratory resistance during PAV.